DATUM PLANE 



1- 



S 2 



5" 3 

 5 M 



14 15 



12 



27 

 26 

 24 29 



21 

 20 



31 



40 



39 



30 



33 



43 



44 



47 



51 



50 



45 



49 



56 

 60 

 57 



54 



63 



66 

 65 



43 



67 68 

 69 



81 



91 



78 

 76 80 



41 



Figure 3.--Distance of bottom of standpipes in gravel bar and bank from established 

 datum plane, Indian Creek, 1958-59. 



slightly higher, and more gradation with depth 

 is apparent than in Latin square 2. 



Variation of Dissolved Oxygen with 

 Depth in the Streambed 



To substantiate results of measurements of 

 dissolved oxygen content of ground water, the 

 variation of dissolved oxygen with depth in 

 the streambed was measured. If the dissolved 

 oxygen content increased with depth or re- 

 mained the same at different depths, then 

 upwelling ground water could be credited as 

 a source of the water of high oxygen content. 

 If, on the other hand, dissolved oxygen con- 

 tent decreased with depth, a shallower source 

 would be indicated. 



Chambers, Allen, and Pressey (1955) have 

 previously reported that the results of tests 

 on streams in Washington State showed that 

 dissolved oxygen of intragravel water de- 

 creased with depth regardless of whether or 

 not salmon had used the gravel for spawning. 



Variation of dissolved oxygen content with 

 depth in streambed gravel was examined in 

 Indian Creek spawning riffles by means of 

 designed experiments. I first used a ran- 

 domized block placement of standpipes, but 

 shifted to 4 by 4 and 3 by 3 Latin squares, 

 because Latin square arrangements could be 

 installed in less time. To define spatial 

 variation several installations were made 

 in different places in the riffle (fig. 2). 



In experiment 1 (fig. 4), 20 standpipes were 

 used, 5 each of which were randomly placed 

 in study area 3 in a 10- by 10-foot square 

 at depths of 5, 10, 15, and 20 inches in the 

 gravel. In 1958 these pipes were sampled on 

 June 3, 5, 7, and 9, and September 13. 



In experiment 2, standpipes were placed in 

 a 4 by 4 Latin square design in study area 1 

 at depths of 5, 10, 15, and 20 inches in the 

 gravel. These pipes were sampled in 1958 on 

 June 4, 5, 7, and 9. 



Experiment 3 was made up of four 3 by 3 

 Latin squares in the same general area as 

 experiment 2. Within each square there were 

 three pipes at 5 inches, three at 10 inches, 

 and three at 20 inches in the gravel. In 1959 

 these pipes were sampled on September 4, 10, 

 and 13. 



Analysis of the random block experiment 

 (experiment 1) shows significant differences in 

 time and with depth (table 3). The 4 by 4 Latin 

 square (experiment 2) revealed a significant 

 difference with depth but no significant dif- 

 ference because of location of pipes (table 4). 

 Analysis of the four 3 by 3 Latin squares 

 (experiment 3) indicates significant differ- 

 ences between depths for squares 3.1, 3.3, 

 and 3.4, but no significant difference with 

 depth in square 3.2 (table 5). Also indicated 

 is a significant difference because of place- 

 ment of pipes in square 3.3. 



